iommu: Get DT/ACPI parsing into the proper probe path

In hindsight, there were some crucial subtleties overlooked when moving
{of,acpi}_dma_configure() to driver probe time to allow waiting for
IOMM

iommu: Get DT/ACPI parsing into the proper probe path

In hindsight, there were some crucial subtleties overlooked when moving
{of,acpi}_dma_configure() to driver probe time to allow waiting for
IOMMU drivers with -EPROBE_DEFER, and these have become an
ever-increasing source of problems. The IOMMU API has some fundamental
assumptions that iommu_probe_device() is called for every device added
to the system, in the order in which they are added. Calling it in a
random order or not at all dependent on driver binding leads to
malformed groups, a potential lack of isolation for devices with no
driver, and all manner of unexpected concurrency and race conditions.
We've attempted to mitigate the latter with point-fix bodges like
iommu_probe_device_lock, but it's a losing battle and the time has come
to bite the bullet and address the true source of the problem instead.

The crux of the matter is that the firmware parsing actually serves two
distinct purposes; one is identifying the IOMMU instance associated with
a device so we can check its availability, the second is actually
telling that instance about the relevant firmware-provided data for the
device. However the latter also depends on the former, and at the time
there was no good place to defer and retry that separately from the
availability check we also wanted for client driver probe.

Nowadays, though, we have a proper notion of multiple IOMMU instances in
the core API itself, and each one gets a chance to probe its own devices
upon registration, so we can finally make that work as intended for
DT/IORT/VIOT platforms too. All we need is for iommu_probe_device() to
be able to run the iommu_fwspec machinery currently buried deep in the
wrong end of {of,acpi}_dma_configure(). Luckily it turns out to be
surprisingly straightforward to bootstrap this transformation by pretty
much just calling the same path twice. At client driver probe time,
dev->driver is obviously set; conversely at device_add(), or a
subsequent bus_iommu_probe(), any device waiting for an IOMMU really
should *not* have a driver already, so we can use that as a condition to
disambiguate the two cases, and avoid recursing back into the IOMMU core
at the wrong times.

Obviously this isn't the nicest thing, but for now it gives us a
functional baseline to then unpick the layers in between without many
more awkward cross-subsystem patches. There are some minor side-effects
like dma_range_map potentially being created earlier, and some debug
prints being repeated, but these aren't significantly detrimental. Let's
make things work first, then deal with making them nice.

With the basic flow finally in the right order again, the next step is
probably turning the bus->dma_configure paths inside-out, since all we
really need from bus code is its notion of which device and input ID(s)
to parse the common firmware properties with...

Acked-by: Bjorn Helgaas <[email protected]> # pci-driver.c
Acked-by: Rob Herring (Arm) <[email protected]> # of/device.c
Signed-off-by: Robin Murphy <[email protected]>
Reviewed-by: Lorenzo Pieralisi <[email protected]>
Reviewed-by: Jason Gunthorpe <[email protected]>
Link: https://lore.kernel.org/r/e3b191e6fd6ca9a1e84c5e5e40044faf97abb874.1740753261.git.robin.murphy@arm.com
Signed-off-by: Joerg Roedel <[email protected]>

show more ...

cdx: Fix possible UAF error in driver_override_show()

Fixed a possible UAF problem in driver_override_show() in drivers/cdx/cdx.c

This function driver_override_show() is part of DEVICE_ATTR_RW, whi

cdx: Fix possible UAF error in driver_override_show()

Fixed a possible UAF problem in driver_override_show() in drivers/cdx/cdx.c

This function driver_override_show() is part of DEVICE_ATTR_RW, which
includes both driver_override_show() and driver_override_store().
These functions can be executed concurrently in sysfs.

The driver_override_store() function uses driver_set_override() to
update the driver_override value, and driver_set_override() internally
locks the device (device_lock(dev)). If driver_override_show() reads
cdx_dev->driver_override without locking, it could potentially access
a freed pointer if driver_override_store() frees the string
concurrently. This could lead to printing a kernel address, which is a
security risk since DEVICE_ATTR can be read by all users.

Additionally, a similar pattern is used in drivers/amba/bus.c, as well
as many other bus drivers, where device_lock() is taken in the show
function, and it has been working without issues.

This potential bug was detected by our experimental static analysis
tool, which analyzes locking APIs and paired functions to identify
data races and atomicity violations.

Fixes: 1f86a00c1159 ("bus/fsl-mc: add support for 'driver_override' in the mc-bus")
Cc: stable <[email protected]>
Signed-off-by: Qiu-ji Chen <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

show more ...

cdx: disable cdx bus from bus shutdown callback

disable cdx bus when bus shutdown is called.

Signed-off-by: Abhijit Gangurde <[email protected]>
Link: https://lore.kernel.org/r/2024120308440

cdx: disable cdx bus from bus shutdown callback

disable cdx bus when bus shutdown is called.

Signed-off-by: Abhijit Gangurde <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

show more ...

module: Convert symbol namespace to string literal

Clean up the existing export namespace code along the same lines of
commit 33def8498fdd ("treewide: Convert macro and uses of __section(foo)
to __s

module: Convert symbol namespace to string literal

Clean up the existing export namespace code along the same lines of
commit 33def8498fdd ("treewide: Convert macro and uses of __section(foo)
to __section("foo")") and for the same reason, it is not desired for the
namespace argument to be a macro expansion itself.

Scripted using

git grep -l -e MODULE_IMPORT_NS -e EXPORT_SYMBOL_NS | while read file;
do
awk -i inplace '
/^#define EXPORT_SYMBOL_NS/ {
gsub(/__stringify\(ns\)/, "ns");
print;
next;
}
/^#define MODULE_IMPORT_NS/ {
gsub(/__stringify\(ns\)/, "ns");
print;
next;
}
/MODULE_IMPORT_NS/ {
$0 = gensub(/MODULE_IMPORT_NS\(([^)]*)\)/, "MODULE_IMPORT_NS(\"\\1\")", "g");
}
/EXPORT_SYMBOL_NS/ {
if ($0 ~ /(EXPORT_SYMBOL_NS[^(]*)\(([^,]+),/) {
if ($0 !~ /(EXPORT_SYMBOL_NS[^(]*)\(([^,]+), ([^)]+)\)/ &&
$0 !~ /(EXPORT_SYMBOL_NS[^(]*)\(\)/ &&
$0 !~ /^my/) {
getline line;
gsub(/[[:space:]]*\\$/, "");
gsub(/[[:space:]]/, "", line);
$0 = $0 " " line;
}

$0 = gensub(/(EXPORT_SYMBOL_NS[^(]*)\(([^,]+), ([^)]+)\)/,
"\\1(\\2, \"\\3\")", "g");
}
}
{ print }' $file;
done

Requested-by: Masahiro Yamada <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Link: https://mail.google.com/mail/u/2/#inbox/FMfcgzQXKWgMmjdFwwdsfgxzKpVHWPlc
Acked-by: Greg KH <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

show more ...

cdx: Fix cdx_mmap_resource() after constifying attr in ->mmap()

Commit 94a20fb9af16 ("sysfs: treewide: constify attribute callback of
bin_attribute::mmap()") missed updating the attr parameter of
cd

cdx: Fix cdx_mmap_resource() after constifying attr in ->mmap()

Commit 94a20fb9af16 ("sysfs: treewide: constify attribute callback of
bin_attribute::mmap()") missed updating the attr parameter of
cdx_mmap_resource(), resulting in a build failure.

drivers/cdx/cdx.c: In function 'cdx_create_res_attr':
drivers/cdx/cdx.c:773:24: error: assignment to 'int (*)(struct file *, struct kobject *, const struct bin_attribute *, struct vm_area_struct *)' from incompatible pointer type 'int (*)(struct file *, struct kobject *, struct bin_attribute *, struct vm_area_struct *)' [-Wincompatible-pointer-types]
773 | res_attr->mmap = cdx_mmap_resource;
| ^

Update cdx_mmap_resource() to match, resolving the build failure.

Fixes: 94a20fb9af16 ("sysfs: treewide: constify attribute callback of bin_attribute::mmap()")
Signed-off-by: Nathan Chancellor <[email protected]>
Tested-by: Thorsten Leemhuis <[email protected]>
Reviewed-by: Thomas Weißschuu <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

show more ...

driver core: have match() callback in struct bus_type take a const *

In the match() callback, the struct device_driver * should not be
changed, so change the function callback to be a const *. This

driver core: have match() callback in struct bus_type take a const *

In the match() callback, the struct device_driver * should not be
changed, so change the function callback to be a const *. This is one
step of many towards making the driver core safe to have struct
device_driver in read-only memory.

Because the match() callback is in all busses, all busses are modified
to handle this properly. This does entail switching some container_of()
calls to container_of_const() to properly handle the constant *.

For some busses, like PCI and USB and HV, the const * is cast away in
the match callback as those busses do want to modify those structures at
this point in time (they have a local lock in the driver structure.)
That will have to be changed in the future if they wish to have their
struct device * in read-only-memory.

Cc: Rafael J. Wysocki <[email protected]>
Reviewed-by: Alex Elder <[email protected]>
Acked-by: Sumit Garg <[email protected]>
Link: https://lore.kernel.org/r/2024070136-wrongdoer-busily-01e8@gregkh
Signed-off-by: Greg Kroah-Hartman <[email protected]>

show more ...

cdx: add MSI support for CDX bus

Add CDX-MSI domain per CDX controller with gic-its domain as
a parent, to support MSI for CDX devices. CDX devices allocate
MSIs from the CDX domain. Also, introduce

cdx: add MSI support for CDX bus

Add CDX-MSI domain per CDX controller with gic-its domain as
a parent, to support MSI for CDX devices. CDX devices allocate
MSIs from the CDX domain. Also, introduce APIs to alloc and free
IRQs for CDX domain.

In CDX subsystem firmware is a controller for all devices and
their configuration. CDX bus controller sends all the write_msi_msg
commands to firmware running on RPU and the firmware interfaces with
actual devices to pass this information to devices

Since, CDX controller is the only way to communicate with the Firmware
for MSI write info, CDX domain per controller required in contrast to
having a CDX domain per device.

Co-developed-by: Nikhil Agarwal <[email protected]>
Signed-off-by: Nikhil Agarwal <[email protected]>
Co-developed-by: Abhijit Gangurde <[email protected]>
Signed-off-by: Abhijit Gangurde <[email protected]>
Signed-off-by: Nipun Gupta <[email protected]>
Reviewed-by: Pieter Jansen van Vuuren <[email protected]>
Reviewed-by: Thomas Gleixner <[email protected]>
Tested-by: Nikhil Agarwal <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

show more ...

cdx: Unlock on error path in rescan_store()

We added locking to this function but these two error paths were
accidentally overlooked.

Fixes: f0af81683466 ("cdx: Introduce lock to protect controller

cdx: Unlock on error path in rescan_store()

We added locking to this function but these two error paths were
accidentally overlooked.

Fixes: f0af81683466 ("cdx: Introduce lock to protect controller ops")
Signed-off-by: Dan Carpenter <[email protected]>
Acked-by: Abhijit Gangurde <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

show more ...

cdx: call of_node_put() on error path

Add a missing call to of_node_put(np) on error.

There was a second error path where "np" was NULL, but that situation is
impossible. The for_each_compatible_n

cdx: call of_node_put() on error path

Add a missing call to of_node_put(np) on error.

There was a second error path where "np" was NULL, but that situation is
impossible. The for_each_compatible_node() loop iterator is always
non-NULL. Just deleted that error path.

Fixes: 54b406e10f03 ("cdx: Remove cdx controller list from cdx bus system")
Signed-off-by: Dan Carpenter <[email protected]>
Acked-by: Abhijit Gangurde <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

show more ...

cdx: Create resource debugfs file for cdx device

resource debugfs file contains host addresses of CDX device resources.
Each line of the resource file describe type of resource, a region
with start-

cdx: Create resource debugfs file for cdx device

resource debugfs file contains host addresses of CDX device resources.
Each line of the resource file describe type of resource, a region
with start-end and flag fields.

Signed-off-by: Abhijit Gangurde <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

show more ...

cdx: create sysfs bin files for cdx resources

Resource binary file contains the content of the memory regions.
These resources<x> devices can be used to mmap the MMIO regions in
the user-space.

Co-

cdx: create sysfs bin files for cdx resources

Resource binary file contains the content of the memory regions.
These resources<x> devices can be used to mmap the MMIO regions in
the user-space.

Co-developed-by: Puneet Gupta <[email protected]>
Signed-off-by: Puneet Gupta <[email protected]>
Signed-off-by: Abhijit Gangurde <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

show more ...

cdx: Explicitly include correct DT includes, again

The DT of_device.h and of_platform.h date back to the separate
of_platform_bus_type before it was merged into the regular platform bus.
As part of

cdx: Explicitly include correct DT includes, again

The DT of_device.h and of_platform.h date back to the separate
of_platform_bus_type before it was merged into the regular platform bus.
As part of that merge prepping Arm DT support 13 years ago, they
"temporarily" include each other. They also include platform_device.h
and of.h. As a result, there's a pretty much random mix of those include
files used throughout the tree. In order to detangle these headers and
replace the implicit includes with struct declarations, users need to
explicitly include the correct includes.

CDX was fixed once, but commit ("cdx: Remove cdx controller list from cdx
bus system") added another occurrence.

Fixes: 54b406e10f03 ("cdx: Remove cdx controller list from cdx bus system")
Signed-off-by: Rob Herring <[email protected]>
Acked-by: Nikhil Agarwal <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

show more ...

cdx: add sysfs for subsystem, class and revision

CDX controller provides subsystem vendor, subsystem device, class and
revision info of the device along with vendor and device ID in native
endian fo

cdx: add sysfs for subsystem, class and revision

CDX controller provides subsystem vendor, subsystem device, class and
revision info of the device along with vendor and device ID in native
endian format. CDX Bus system uses this information to bind the cdx
device to the cdx device driver.

Co-developed-by: Puneet Gupta <[email protected]>
Signed-off-by: Puneet Gupta <[email protected]>
Co-developed-by: Nipun Gupta <[email protected]>
Signed-off-by: Nipun Gupta <[email protected]>
Signed-off-by: Abhijit Gangurde <[email protected]>
Reviewed-by: Pieter Jansen van Vuuren <[email protected]>
Tested-by: Nikhil Agarwal <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

show more ...

cdx: add sysfs for bus reset

Add sysfs interface reset to reset all the devices on the CDX bus.

Signed-off-by: Abhijit Gangurde <[email protected]>
Link: https://lore.kernel.org/r/2023101716

cdx: add sysfs for bus reset

Add sysfs interface reset to reset all the devices on the CDX bus.

Signed-off-by: Abhijit Gangurde <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

show more ...

cdx: add support for bus enable and disable

CDX bus needs to be disabled before updating/writing devices
in the FPGA. Once the devices are written, the bus shall be
rescanned. This change provides s

cdx: add support for bus enable and disable

CDX bus needs to be disabled before updating/writing devices
in the FPGA. Once the devices are written, the bus shall be
rescanned. This change provides sysfs entry to enable/disable the
CDX bus.

Co-developed-by: Nipun Gupta <[email protected]>
Signed-off-by: Nipun Gupta <[email protected]>
Signed-off-by: Abhijit Gangurde <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

show more ...

cdx: Register cdx bus as a device on cdx subsystem

While scanning for CDX devices, register newly discovered bus as a
cdx device. CDX device attributes are visible based on device type.

Signed-off-

cdx: Register cdx bus as a device on cdx subsystem

While scanning for CDX devices, register newly discovered bus as a
cdx device. CDX device attributes are visible based on device type.

Signed-off-by: Abhijit Gangurde <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

show more ...

cdx: Create symbol namespaces for cdx subsystem

Create CDX_BUS and CDX_BUS_CONTROLLER symbol namespace for cdx bus
subsystem. CDX controller modules are required to import symbols from
CDX_BUS_CONTR

cdx: Create symbol namespaces for cdx subsystem

Create CDX_BUS and CDX_BUS_CONTROLLER symbol namespace for cdx bus
subsystem. CDX controller modules are required to import symbols from
CDX_BUS_CONTROLLER namespace and other than controller modules to
import from CDX_BUS namespace.

Signed-off-by: Abhijit Gangurde <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

show more ...

cdx: Introduce lock to protect controller ops

Add a mutex lock to prevent race between controller ops initiated by
the bus subsystem and the controller registration/unregistration.

Signed-off-by: A

cdx: Introduce lock to protect controller ops

Add a mutex lock to prevent race between controller ops initiated by
the bus subsystem and the controller registration/unregistration.

Signed-off-by: Abhijit Gangurde <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

show more ...

cdx: Remove cdx controller list from cdx bus system

Remove xarray list of cdx controller. Instead, use platform bus
to locate the cdx controller using compat string used by cdx
controller platform d

cdx: Remove cdx controller list from cdx bus system

Remove xarray list of cdx controller. Instead, use platform bus
to locate the cdx controller using compat string used by cdx
controller platform driver.
Also, use ida to allocate a unique id for the controller.

Signed-off-by: Abhijit Gangurde <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

show more ...

cdx: add support for bus mastering

Introduce cdx_set_master() and cdx_clear_master() APIs to support
enable and disable of bus mastering. Drivers need to use these APIs to
enable/disable DMAs from t

cdx: add support for bus mastering

Introduce cdx_set_master() and cdx_clear_master() APIs to support
enable and disable of bus mastering. Drivers need to use these APIs to
enable/disable DMAs from the CDX devices.

Signed-off-by: Nipun Gupta <[email protected]>
Reviewed-by: Pieter Jansen van Vuuren <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Alex Williamson <[email protected]>

show more ...

cdx: fix driver managed dma support

The devices on cdx could be bound to drivers with the device
DMA managed by kernel drivers or user-space applications.
As multiple devices can be placed in the sa

cdx: fix driver managed dma support

The devices on cdx could be bound to drivers with the device
DMA managed by kernel drivers or user-space applications.
As multiple devices can be placed in the same IOMMU group, the
DMA on these devices must either be entirely under kernel control
or userspace control. Fix the CDX bus driver to acknowlege the
driver_managed_dma flag and call the appropriate iommu APIs.

Fixes: 2959ab247061 ("cdx: add the cdx bus driver")
Signed-off-by: Nipun Gupta <[email protected]>
Reported-by: Alex Williamson <[email protected]>
Closes: https://lore.kernel.org/lkml/[email protected]/
Reviewed-by: Nikhil Agarwal <[email protected]>
Message-ID: <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

show more ...

cdx: fix build failure due to sysfs 'bus_type' argument needing to be const

Commit 75cff725d956 ("driver core: bus: mark the struct bus_type for
sysfs callbacks as constant") missed at least one cas

cdx: fix build failure due to sysfs 'bus_type' argument needing to be const

Commit 75cff725d956 ("driver core: bus: mark the struct bus_type for
sysfs callbacks as constant") missed at least one case - the CDX bus
driver. Probably because Greg didn't notice the build failure, because
it only ends up being enabled on arm64.

And I missed it during the merge, because while I do arm64 builds these
days, I don't do them in between each pull. So it took a while for me
to notice the breakage, rather than me just fixing it in the driver core
merge that brought this failure case in.

Maybe we should remove the CDX_BUS dependency on arm64 when COMPILE_TEST
is on?

Cc: Greg Kroah-Hartman <[email protected]>
Cc: Nipun Gupta <[email protected]>
Cc: Nikhil Agarwal <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

show more ...

cdx: add device attributes

Create sysfs entry for CDX devices.

Sysfs entries provided in each of the CDX device detected by
the CDX controller
- vendor id
- device id
- remove
- reset of the de

cdx: add device attributes

Create sysfs entry for CDX devices.

Sysfs entries provided in each of the CDX device detected by
the CDX controller
- vendor id
- device id
- remove
- reset of the device.
- driver override

Signed-off-by: Puneet Gupta <[email protected]>
Signed-off-by: Nipun Gupta <[email protected]>
Signed-off-by: Tarak Reddy <[email protected]>
Reviewed-by: Pieter Jansen van Vuuren <[email protected]>
Tested-by: Nikhil Agarwal <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

show more ...

cdx: add the cdx bus driver

Introduce AMD CDX bus, which provides a mechanism for scanning
and probing CDX devices. These devices are memory mapped on
system bus for Application Processors(APUs).

C

cdx: add the cdx bus driver

Introduce AMD CDX bus, which provides a mechanism for scanning
and probing CDX devices. These devices are memory mapped on
system bus for Application Processors(APUs).

CDX devices can be changed dynamically in the Fabric and CDX
bus interacts with CDX controller to rescan the bus and
rediscover the devices.

Signed-off-by: Nipun Gupta <[email protected]>
Reviewed-by: Pieter Jansen van Vuuren <[email protected]>
Tested-by: Nikhil Agarwal <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>

show more ...